Through a series of three protocols, we are using functional magnetic resonance imaging (fMRI) and magnetic resonance imaging spectroscopy (MRS) to examine the effects of pharmacologic treatment on neurocognitive and neurochemical correlates of pediatric mood and anxiety disorders. In one aspect of this project, we are attempting to document deficits in brain systems mediating reward-related processes, attention bias, and emotional memory in pediatric mood and anxiety disorders. In a second aspect of this project, we are examining the effects of pharmacological agents on symptoms of mood and anxiety disorders, including both acute episodes as well as treatment resistant symptoms. In a third aspect of this project, we are examining the effects of treatment on brain activation using fMRI and neurochemistry using MRS. For the acute treatment study, total of 155 children, adolescents and adults will be recruited, including 45 juveniles with a current anxiety disorder, 60 with current major depression, 25 adolescents with no psychiatric disorder. Additionally, 25 psychiatrically healthy adults will be examined. Before the initiation of treatment, these 155 subjects will be tested using behavioral and fMRI paradigms designed to examine brain regions engaged by attending to and encoding emotionally salient stimuli. After this initial fMRI test, the 105 subjects with depression or an anxiety disorder will be randomized to receive eight weeks of double-blind treatment with fluoxetine or placebo. All 155 subjects will be re-tested after eight-weeks using the same two fMRI paradigms. For the study of treatment resistance, 60 subjects will be treated either with an SSRI plus placebo or an SSRI plus lithium. Another 30 subjects will be treated openly with an SSRI. We will acquire MRS measures of neurochemical compounds in these subjects before and after treatment. Prior neuropsychological studies in children as well as in adults note that mood and anxiety disorders are associated with deficits in attention modulation and emotional memory. Prior imaging studies in healthy adults note that tasks requiring attention modulation or emotional memory engage cortico-limbic brain regions previously implicated in adult mood and anxiety disorders. These regions include the amygdala, ventro-medial prefrontal cortex, cingulate gyrus, and hippocampus. As a result, we hypothesize that fMRI attention modulation and emotional memory paradigms will engage these cortico-limbic brain regions in both psychiatrically healthy and impaired subjects. However, these groups will differ in the degree of cortico-limbic engagement. During the first year of this study, we successfully enrolled 11 patients in this protocol. Each of these subjects received neurocognitive examinations, and a subset received fMRI examinations. As part of our studies in healthy subjects, we also successfully developed each of the fMRI protocols that will be used in the current project. We are currently just beginning to enter subjects into the aspects of this project that examine treatment resistant depression.